PROJECT SUMMARY Gastric cancer (GC) is the third worldwide cause of cancer-related death and a leading cause of cancer related mortality and morbidity in U.S. Hispanics. Precision GC medicine lags behind most other cancers and there is an urgent need to further understand its etiology. Addressing existing research gaps will facilitate precision prevention, treatment, and will improve survival outcomes and disparities. In a recent study, we identified germline mutations in the homologous recombination (HR) repair genes PALB2, BRCA1 and RAD51C in multiple GC patients, suggesting that the HR pathway is important in GC risk. Tumors from these patients have a HR-deficient (HRD) mutational signature, a signature that has been reported in ~7% of sporadic non- Hispanic white (NHW) tumors. Interestingly, our unpublished data in Hispanics suggest a higher prevalence of gastric tumors with the HRD phenotype (19%). These recent findings have dual importance in precision GC medicine as HR gene testing can now be considered in prevention through germline HR gene mutation testing and patients with HRD tumors may benefit from PARP inhibitors, providing a promising option to the only other molecularly-guided therapies available for GC treatment. The long-term goal of our program is to generate knowledge for improving GC outcomes and disparities. The objectives of this application are to identify new GC genes and to characterize the genomic, clinico-pathological and risk profile of Hispanic gastric tumors and of gastric HRD tumors in particular. Our hypotheses are that the HR pathway is enriched with GC risk variants and that Hispanic and HRD GCs have unique genomic profiles. In Aim 1, we will identify germline mutations in 384 familial and early-onset patients of Hispanic origin using targeted sequencing of all HR pathway genes. Our analyses will focus on identifying the main clinical characteristics of germline HR gene mutation carriers and on estimating HR gene mutation prevalence and penetrance. In Aim 2, we will conduct exome sequencing and copy number analyses of paired tumor/normal samples from 300 Hispanic GC patients that were recruited in a multi-center study in Latin America. The genomic data will be used to identify novel GC drivers that may be population-specific, to estimate the prevalence of known GC molecular subtypes and driver genes previously identified in NHWs and to carry out comparisons between molecular phenotypes and the extensive clinical and risk factor data available in all these samples. In Aim 3, we will compile HRD data from published studies and from the present study to identify the main clinical, histological and risk factor characteristics of this ?druggable? tumor type. The proposed study builds on extensive pilot data of newly-discovered HR pathway GC genes. We will capitalize on our excellent and well-characterized patient resources and samples, which include a high-risk Hispanic population. This study focuses on a significant cancer disparity and we will generate data essential for understanding GC genomics and etiology and for the development of molecularly-guided therapies.